another 2p... I agree with Les about the 'clean' interface. However, IMHO, it would be a loss to get rid of it all together and just have examples with the jar task in the doc.
The war allows to clearly specify the important elements that should be in it. It gives special treatements to files that matter such as the webxml file and performs some degree of checks. One enhancement that one could possibly add thanks to this approach is a check on the validity of the webxml file with sun's dtd. Also the build file becomes then far more readable. I am using it all the time for my development. I found it so useful that I actually wrote an <ear> task myself very, very similar to Les'. I also wrote one for client apps and a version of an <ejbjar> task with attributes closer to <war>. With all that, it is far easier to know where one stands and my build file is shorter and easier to read. Also there was discussion about adding a <metainf> nested element in the <jar> task (see the "[SUBMIT] Ear.java task" thread), which IMHO is a good approach. Any thoughts? Emmanuel ----- Original Message ----- From: "Les Hughes" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Tuesday, April 03, 2001 12:29 PM Subject: RE: continuing development for ant 1.4 > > > Just my 2p..... > > Hmm, removing WAR ? I know there's been some discussion about the use of > zipfilesets but doesn't this approach make things less easy for end users? > Personally, I prefer the 'clean' syntax of <war> but I guess as long as > there's a war, ear, rar, car example in the docs under jar then I suppose it > doesn't matter. > > I'll resend ear in a mo. > > > Bye, > > Les > > > > -----Original Message----- > > From: Peter Donald [mailto:[EMAIL PROTECTED] > > Sent: 03 April 2001 10:59 > > To: [EMAIL PROTECTED] > > Subject: RE: continuing development for ant 1.4 > > > > > > At 10:33 3/4/01 +0100, Les Hughes wrote: > > >It was pretty obvious that you guys had other things on your > > minds recently > > >but when can we expect to see new 1.x tasks making it into > > CVS - once all of > > >the discussion is over maybe? > > > > as soon as something gets enough votes. I only personally use a small > > subset of ant tasks and thus don't really feel confident > > about checking in > > changes without someone else checking it. Whenever I ask for > > other people > > to check it no one does ... hence it doesn't get added ;) > > > > >In my case, I'm waiting on a decision on <ear> > > > > If you want it then I will add it in but there was concern > > that it is no > > longer needed. Resend it and if no one -1s it then I will check it in. > > > > >and more importantly, Kirk's <p4counter>. > > > > no idea about that ;) > > > > >Also, I was wondering will 'old' 1.x tasks will need to be > > ported to 2.x > > >sometime in the not to distant future? > > > > I would more say "distant" future as we haven't even got into > > design phase > > of ant2 so ... ;) > > > > >I'd like to start thinking about > > >cleaning up the P4 stuff, in particular around the use (or > > not) of Oro for > > >regexp matching in p4base but I'd rather wait for the 2.x > > framework to be a > > >little firmer before starting if it'll require task re-writes. > > > > rewrite it now if you want ;) > > Cheers, > > > > Pete > > > > *-----------------------------------------------------* > > | "Faced with the choice between changing one's mind, | > > | and proving that there is no need to do so - almost | > > | everyone gets busy on the proof." | > > | - John Kenneth Galbraith | > > *-----------------------------------------------------* > >
